US 6215170 B1
The device described permits selection between two design options of an integrated circuit by causing a corresponding circuit unit of the integrated circuit to adopt one of two possible different operative states. It comprises an inverter, of which the output terminal is connected to the control terminal of the circuit unit and the input terminal is connected to a first supply terminal by means of a conductor which can be broken by means outside the integrated circuit, and to the second supply terminal by means of a capacitor in parallel with a diode connected for reverse conduction. The device described does not require control signals, takes up a very small area, has practically zero consumption, and can be formed in unlimited numbers on the same integrated circuit.
1. An integrated circuit comprising:
first and second supply terminals coupled to a substrate;
a plurality of semiconductor layers formed on said substrate and having a set of integrated circuit components formed therein;
a circuit unit of said set of components operable to be configured in one of a number of configurations in response to a signal;
a removable link of said set of components having a first link terminal coupled to said first supply terminal, and a second link terminal;
a buffer of said set of components having a buffer input terminal coupled to said second link terminal and a buffer output terminal coupled to said circuit unit, said buffer operable to generate said signal at said buffer output terminal;
an insulator on a portion of said substrate being patterned to expose first and second direct substrate contacts, said first contact being coupled to said second link terminal and said second contact being coupled to said second supply terminal formed on said insulator between said first and second contacts;
a first region of a first conductivity type in a surface area of said substrate under said first contact; and
a second region of a second conductivity type formed under said insulator and extending past said second contact, said second region is laterally spaced from said first region, and said first and second regions and insulator form a capacitive element in parallel with a diode within said integrated circuit.
2. The integrated circuit of claim 1 wherein the charge value of said capacitive element is less than a threshold voltage value of said buffer.
3. The integrated circuit of claim 1 wherein said second link terminal has a high impedance when a fuse is blown.
4. The circuit of claim 1 wherein said circuit unit and said buffer comprise components formed in said substrate.
5. The circuit of claim 1 wherein said first and second regions are contained within a well region of said second conductivity.
6. The circuit of claim 5 wherein said substrate is of said first conductivity type.
7. The circuit of claim 1 wherein said substrate is of said second conductivity type.
8. The circuit of claim 1 wherein a power consumption is reduced.
This application is a continuation of U.S. patent application Ser. No. 08/778,416, filed Dec. 31, 1996 now abandoned, entitled “DEVICE FOR SELECTING DESIGN OPTIONS IN AN INTEGRATED CIRCUIT,” which is pending, incorporated herein by reference.
The present invention relates to the design of integrated semiconductor circuits and, more particularly, to a parallel capacitor and diode structure used in a device for selecting design options in an integrated circuit.
It is known that, during the design of a complex integrated circuit, it is often difficult to establish with certainty beforehand the most advantageous circuit configuration or the most suitable component to achieve a predetermined functional effect. For example, it may be difficult, if not impossible, to evaluate the best time constant for a certain function, or the most appropriate intervention threshold for a given circuit, or the ideal current which a current-generator should supply. Moreover, it may also be useful to test a section of the integrated circuit in several different operative conditions, for example, with and without a network for correcting a parameter, in order to evaluate the effect of the correction, as in the case of an “offset”-canceling circuit, the exclusion of which would allow the magnitude of the “offset” to be evaluated.
It is also known that, during the design of an integrated circuit, it may be convenient to leave the selection between several variants of the circuit configurations and/or parameters open, to permit the production of integrated circuits the general functional characteristics of which are substantially identical, but which satisfy slightly different applicational requirements.
Thus, to allow for the “design options” described above, it is often necessary, at the design stage, to provide for modifiable connections or for selectively addressable memory cells or registers which enable the inputs of associated circuit units to be modified so as to adopt different operative states in dependence on the states of the aforesaid modifiable connections or of the memory cells or registers.
The various options are then tested and, of these, the one which offers the functional characteristics closest to those desired is selected and the final topography of the integrated circuit to be put into production is consequently defined.
The known techniques for satisfying the design requirements described above provide for the use of terminals of the device by means of which connection may or may not be established, of fuses within the integrated circuit which can be broken by means of high currents applied between suitable terminals of the integrated circuit, of EPROM cells, of programmable register cells, or of different metallization masks for the various design options. All of these techniques, except for the last mentioned, take up quite large areas of the integrated circuit; in particular, the use of terminals of the device requires the formation of contact pads which, as is known, necessarily have large dimensions; fuses require wide metal connecting strips because of the high currents necessary to break them and are also unreliable; in addition to the space which they take up, EPROM cells also require a special manufacturing process. In any case, for all these techniques, it is necessary to provide piloting signals which, in order to control the integrated circuit, naturally have to be conducted to the corresponding points by means of suitable metal tracks which are also very bulky.
As is known to a designer of integrated circuits, after the most favorable design options have been defined, modification of the topography of the integrated circuit in order to recover the areas intended for the test components and/or connections often cannot be accomplished, and these areas are therefore inevitably lost, detracting from the compactness of the final integrated structure.
As regards the technique which provides for different metallization masks for the various options, that is, variations in the mask for the formation of the last layer of connections in the sequence of steps of the manufacturing process, the problem is not the area taken up since, in this case, additional components or connections are not necessary, but, rather, the cost and the time necessary to put it into practice. In fact, each variant to be tested corresponds to a different mask, which itself is quite expensive, and to the manufacture of a batch of test chips, which requires a very long waiting time to evaluate the test. This technique is therefore suitable only when the number of variants to be tested is very low.
An object of the present invention is to propose a device for selecting between design options in an integrated circuit which is reliable, which has practically zero power consumption, which takes up a very small area, which does not require additional masks or process steps, and which is usable without limitations on the number of variants to be tested.
Another object of the invention is to provide a structure having a diode and capacitor coupled in parallel that can be implemented in standard bipolar technology as well as in CMOS technology containing an N or P well.
According to one aspect of the present invention, an integrated circuit is provided for selecting one of a number of design configurations for an integrated device. The circuit includes first and second supply terminals and a substrate having a first conductivity. A plurality of semiconductor layers are formed on said substrate and have a set of integrated circuit components formed therein. A circuit unit of the set of components is operable to be configured in one of a number of configurations in response to a signal. A removable link of the set of components has a first link terminal coupled to the first supply terminal. A buffer of the set of components has a buffer input terminal coupled to a second link terminal and a buffer output terminal coupled to the circuit unit. The buffer is operable to generate said signal at said buffer output terminal. An insulator is provided over a portion of said substrate being patterned to expose first and second direct substrate contacts. The first contact is coupled to the second link terminal and the second contact is coupled to the second supply terminal formed on the insulator between the first and second contacts. A first region of the first conductivity type in a surface area of the substrate is formed under the first contact. A second region of a second conductivity type is formed under the insulator and extends past the second contact. The second region is laterally spaced from the first region, and the first and second regions and insulator form a capacitive element in parallel with a diode within the integrated circuit.
The invention will be better understood from the following description, which is given by way of nonlimiting example, with reference to the appended drawings.
FIG. 1 is a schematic circuit diagram showing an integrated circuit comprising a device according to the invention.
FIG. 2 is a schematic circuit diagram showing a variant of the device according to the invention.
FIG. 3 is a circuit diagram of the device of FIG. 1.
FIG. 4 shows a cross-section of a portion of an integrated circuit having a device according to a first embodiment of the invention.
FIG. 5 shows a cross-section of a portion of an integrated circuit in FIG. 1 according to a second embodiment of the invention.
FIG. 6 shows a cross-section of a portion of the integrated circuit in FIG. 1 according to a third embodiment of the invention.
FIG. 7 shows a cross-section of a portion of the integrated circuit as illustrated in FIG. 2 according to a fourth embodiment.
In FIG. 1, an integrated circuit 10 comprises a device 11 according to the invention, and a circuit unit 12. Unit 12 represents, by way of example, a very simple form of a circuit which can adopt two different operative states in dependence on a signal applied to a control terminal thereof. Unit 12 comprises two resistors R1 and R2 connected in series, and an electronic switch shown as a switch SW, which is connected in parallel with R2 and has a control terminal 13. Clearly, in this case, the two different operative states are two different values of the resistance between the terminals 14 and 15 of the circuit unit 12, depending upon the open or closed state of the electronic switch SW.
The selection device 11 comprises an inverter 16, of which the output 17 is connected to the control terminal 13 of the electronic switch SW, and the input terminal 18 is connected to one of the supply terminals of the integrated circuit, indicated Vdd, by means of a conductor which is breakable between two points A and B. The input 18 of the inverter 16 is also connected to the other supply terminal of the integrated circuit, indicated by the ground symbol, by capacitive means which, in this embodiment, is a capacitor C1 connected in parallel with a component with unidirectional conduction which, in this embodiment, is a diode D1 connected for reverse conduction as shown.
The connection between the input 18 and the supply terminal Vdd, or at least the portion thereof between the points A and B, comprises a conductive track or link, preferably of metal, which extends on the integrated circuit or, more precisely, on the surface of the “chip” of semiconductor material, on which it is formed, and can easily be recognized by the usual instruments for inspecting the surfaces of semiconductor devices and can be broken, removed, or opened by means of a normal mechanical or laser cutting instrument.
With the connection between A and B intact, the input of the inverter 15 is at a high level and its output 17 is therefore at a low level. Under these conditions, the electronic switch SW is not activated, that is, it is open and the state of the unit 12 between its terminals 14 and 15 is defined by the sum of the resistances of the two resistors R1 and R2 in series.
If the connection between the points A and B is broken, the potential at the input terminal 18 is substantially that of ground, i.e., zero volts, and the output of the inverter 16 is therefore at a high level. Under these conditions, the electronic switch SW is activated, that is, it is closed, and the state of the unit 12 between its terminals 14 and 15 is defined by the resistance of the resistor R1 alone, since the resistor R2 is short-circuited by the switch SW. The function of the diode D1 is to prevent the input terminal 18, the resistance of which with respect to ground is practically infinite, from being charged to a positive potential; this potential could reach the threshold at which the inverter is triggered as a result of any charges trapped in the dielectric of the capacitor C1 and in the input capacitance of the inverter 16 after long periods of operation of the integrated circuit. The diode D1 introduces a finite, although very high, resistance between the input terminal 18 and ground. That is, diode D1 allows a leakage current to flow between terminal 18 and ground when charge accumulates at node 18. This leakage current discharges node 18, and thus prevents the accumulation of charge at node 18 when the connection between points A and B is opened or otherwise removed. On the other hand, diode D1 does not affect the operation of the device when the connection between the points A and B is intact, since it is connected for reverse conduction.
FIG. 2 shows the inverse version of the device 11 of FIG. 1. In this case, in fact, the output terminal 17′ of the inverter 16′ is at a high level if the connection between its input 18′ and ground is intact and is at a low level if the connection is broken. The capacitor C2 and the diode D2 have the same functions as the capacitor C1 and the diode D1 of FIG. 1; they serve, respectively, to ensure that, after the connection between A′ and B′ has been cut, the input terminal 18′ of the inverter 16′ remains at a high value, that is, close to the positive supply voltage Vdd, and to constitute a finite resistance between that terminal and the terminal Vdd.
In the practical embodiment of the device according to the invention, shown in FIG. 3, the inverter 16 comprises a pair of complementary field-effect transistors, that is, a p-channel transistor, indicated MP, and an n-channel transistor, indicated MN, having common drain electrodes, forming the output terminal 13, common gate electrodes, forming the input terminal 18, and source electrodes, of which one is connected to the positive supply terminal Vdd and the other to the negative supply or ground.
The diagram also shows the intrinsic capacitances CMP, CMN, existing between the gate and source terminals of the transistors MP and MN, respectively, as well as a capacitor C3 and a diode D3 which correspond, respectively, to the capacitor C1 and to the diode D1 of FIG. 1. In selecting the capacitance of the capacitor C3, account must be taken of the intrinsic capacitances indicated above. It should be noted, in this connection, that the capacitance CMN increases the effect of the capacitor C3 between the input terminal 18 and ground, whereas the capacitance CMP tends to reduce that effect. If CMN, CMP and C3 indicate the capacitances of the capacitors indicated by the same symbols in the drawing and Vs indicates the threshold of the inverter, that is, the voltage between the input terminal 18 and ground which causes the logic state at the output terminal 13 to change, the capacitance of the capacitor C3 must satisfy the following relationship:
The transistors MP and MN can be of a size such that the threshold is sufficiently high and the transistor MP can be very small and hence with a low capacitance CMP, so that the capacitor C3 can be very small.
The device described above can be produced particularly advantageously by a method of manufacturing CMOS silicon integrated circuits in which the capacitors can be formed with the use, as electrodes, of a strongly doped region of the silicon substrate, for example, a region implanted with arsenic on a p-type substrate, and of an opposed portion of polycrystalline silicon, separated by a thin layer of silicon dioxide. In this case, the capacitor C3 is formed as shown in FIG. 4, in which the silicon substrate is indicated 20, the strongly doped n-type region which forms the first electrode is indicated 21, the silicon dioxide which forms the dielectric is indicated 22, and the layer of polycrystalline silicon which forms the second electrode is indicated 23. A metal connection 24 a formed on the surface of the region 21 leads to the gate electrodes of the two transistors (not shown), that is, to the input terminal 18 of the inverter 16, and the second electrode 23 is connected to the ground terminal, and also to the substrate 20 at p+region 25 by metal region 24 b.
FIG. 5 shows a second embodiment of the parallel coupled diode D1 and capacitor C1 structure as illustrated in FIG. 4. With this structure, after the first dielectric 22 is formed on the substrate 20 over the implanted region 21, the dielectric is patterned to expose two substrate contacts. Next, an implantation step is performed, typically using Boron dopant ions, to provide a strongly doped p-type region in the substrate. The strongly doped p-type region is formed under the second contact to establish the second electrode for the diode D1. Subsequently, a polycrystalline silicon layer 23 is deposited over the dielectric 22 and into the second contact coupling the second electrode to the polycrystalline silicon layer 23. The polycrystalline silicon is doped with Boron either during deposition or by a separate doping step after deposited. Metal connection 24 is formed in the first contact. This structure combines the two ground leads of the embodiment of FIG. 4 into one single lead. Consequently, the diode, as illustrated by a dashed line, and a capacitor are formed in parallel between the input terminal 18 and the ground terminal 26.
The circuit structures of FIGS. 1 and 2 can be implemented in any complementary bipolar technology and in any CMOS technology by adding an isolation well 30 to the structure of FIG. 5 as illustrated by FIGS. 6 and 7. The isolation wells 30 have a first and second conductivity type, respectively, and are formed by typical implantation and diffusion steps. For the circuit of FIG. 1, a P-type dopant, typically boron ions, are implanted to form the isolation well 30 structure as illustrated in FIG. 6. On the other hand, as illustrated in FIG. 7, an N-type dopant, typically Arsenic ions, are implanted to form the isolation well 30 for the circuit of FIG. 2.
It should be noted that the diode D3 of FIG. 3 is formed together with the capacitor C3, without the use of additional silicon area and without further process steps. As can easily be seen, the diode is in fact formed by the junction between the region 21 and the substrate 20. Similarly, with a different process, it would be possible to produce a diode the cathode of which had a certain capacitance with respect to ground, so that it would not be necessary to form an actual capacitor, since the parasitic or inherent capacitance of the diode could be used to perform the function of the capacitor C3.
As can easily be confirmed, the device according to the invention can be formed in a very small area, since it is formed solely by two signal transistors, that is, very low-power transistors, by a capacitor of small capacitance, and by a diode which, as in the embodiment described above, results intrinsically from the structure of the capacitor and, moreover, requires no control signal and therefore no additional conductive track on the integrated circuit. Furthermore, it has practically zero consumption since the currents involved are only leakage currents. In a practical embodiment on an integrated circuit with an area of 25 mm2, the device according to the invention occupied an area of about 500 μm2.
By way of indication, with one of the known techniques cited at the beginning, except for that which provides for different metallization masks for the different options, an area at least 10 times greater would be taken up in order to obtain similar design options.
Naturally, many of these devices may be formed at the same time on the same integrated circuit so as to enable all the possible design options to be evaluated. Once all the most favorable design variants have been identified, a single mask can be provided to define the connection layout of the integrated circuit intended for production. Any future variations will be possible by the modification solely of this mask.
Although only four embodiment of the invention has been described and illustrated, clearly many variants are possible within the spirit and scope of the same innovative concept. For example, the inverter may be replaced by non-complementary field-effect transistors, or by complementary or non-complementary bipolar transistors, or by a noninverting buffer, according to the type of integrated circuit for which the device according to the invention is intended, and according to the manufacturing process available at the time in question.